Kinetics of the Autoreduction of Hexavalent Americium in Aqueous

Jun 29, 2017 - School of Chemistry, The University of Manchester, Oxford Road, Manchester M13 9PL, United Kingdom. Inorg. Chem. , 2017, 56 (14), ...
1 downloads 0 Views 1MB Size
Download PDF
Article pubs.acs.org/IC

Kinetics of the Autoreduction of Hexavalent Americium in Aqueous Nitric Acid Travis S. Grimes,*,† Gregory P. Horne,‡,§ Christopher J. Dares,∥ Simon M. Pimblott,⊥,# Stephen P. Mezyk,‡ and Bruce J. Mincher† †

Idaho National Laboratory, P.O. Box 1625, Idaho Falls, Idaho 83415, United States California State University at Long Beach, Long Beach, California 90804, United States § Radiation Research Laboratory, University of Notre Dame, Notre Dame, Indiana 46556, United States ∥ Florida International University, Miami, Florida 33199, United States ⊥ Dalton Cumbrian Facility, Westlakes Science and Technology Park, The University of Manchester, Cumbria CA24 3HA, United Kingdom # School of Chemistry, The University of Manchester, Oxford Road, Manchester M13 9PL, United Kingdom ‡

S Supporting Information *

ABSTRACT: The rate of reduction of hexavalent 243Am due to self-radiolysis was measured across a range of total americium and nitric acid concentrations. These so-called autoreduction rates exhibited zero-order kinetics with respect to the concentration of hexavalent americium, and pseudo-first-order kinetics with respect to the total concentration of americium. However, the rate constants did vary with nitric acid concentration, resulting in values of 0.0048 ± 0.0003, 0.0075 ± 0.0005, and 0.0054 ± 0.0003 h−1 for 1.0, 3.0, and 6.5 M HNO3, respectively. This indicates that reduction is due to reaction of hexavalent americium with the radiolysis products of total americium decay. The concentration changes of Am(III), Am(V), and Am(VI) were determined by UV−vis spectroscopy. The Am(III) molar extinction coefficients are known; however, the unknown values for the Am(V) and Am(VI) absorbances across the studied range of nitric acid concentrations were determined by sensitivity analysis in which a mass balance with the known total americium concentration was obtained. The new extinction coefficients and reduction rate constants have been tabulated here. Multiscale radiation chemical modeling using a reaction set with both known and optimized rate coefficients was employed to achieve excellent agreement with the experimental results, and indicates that radiolytically produced nitrous acid from nitric acid radiolysis and hydrogen peroxide from water radiolysis are the important reducing agents. Since these species also react with each other, modeling indicated that the highest concentrations of these species available for Am(VI) reduction occurred at 3.0 M HNO3. This is in agreement with the empirical finding that the highest rate constant for autoreduction occurred at the intermediate acid concentration.

1. INTRODUCTION Changes in the electronic structure of metal ions can cause profound changes in their chemical properties. For example, the hydrometallurgical industry exploits these chemical-property changes through the reduction of Cu2+, Ni2+, Zn2+, Cd2+, and Al3+ ions in solution and electrodeposition of the zerovalent metals onto insoluble cathodes.1−3 In acidic aqueous solutions, the 4f-element lanthanide metal ions prefer the trivalent oxidation state. Actinide elements exhibit a variety of oxidation-state preferences due to the diminished shielding of the outer 5f electrons, resulting in slightly more diffuse valence electron orbitals. This difference is manifest with high oxidation states (IV−VI) being typical for the early actinides. Actinides able to obtain the highest valence states (i.e., V or VI) hydrolyze to form linear dioxo cations in aqueous solution; in this process, the shape of the ion changes from a spherical geometry to a linear geometry with two axial oxygen atoms © 2017 American Chemical Society

bonded to the metal ion in a trans configuration, in turn altering their charge density.4−6 The preferred oxidation state of americium is trivalent in acidic solution. However, the hexavalent state is possible, but is a challenge to reach given the intermediate high Am(IV/III) redox couple (2.62 V versus the saturated calomel electrode, SCE), and also a challenge to maintain since it is thermodynamically and radiolytically unstable.7 Understanding the mechanisms which result in its redox instability are paramount to the design of new experiments to further explore the properties of this relatively unknown species. In addition to enriching our knowledge of 5f-element chemistry, the hexavalent state of americium may have practical utility toward nuclear fuel recycle separations. Metal-ion charge Received: April 18, 2017 Published: June 29, 2017 8295

DOI: 10.1021/acs.inorgchem.7b00990 Inorg. Chem. 2017, 56, 8295−8301

Article

Inorganic Chemistry

and americium solutions of 1.0 and 3.0 M HNO3 required 24 h of vigorous shaking to yield ≈80−85% Am(VI), with smaller amounts of Am(V) and residual Am(III). After being shaken, the heterogeneous Am/NaBiO3 mixture was transferred to a 10 mL syringe with a glass Pasteur pipet and the undissolved NaBiO3 filtered off using a 25 mm syringe disk filter with a 0.2 μm PTFE membrane (PALL Life Sciences, VWR). The solution containing the oxidized americium was collected in a 20 mL glass scintillation vial. The americium solution was then transferred into a screw-cap semimicro 10 mm path-length quartz cuvette (Starna) for UV−vis analysis. A single experiment was conducted in 3.0 M HNO3 with 20 mg mL−1 copper(III) periodate (Na5[Cu(IO5)(OH)2]·12H2O) as the oxidizing agent. Unlike sodium bismuthate, copper(III) periodate is immediately soluble, and the solution was transferred without filtration to a 10 mm cuvette for UV−vis analysis. 2.3. Data Collection and Interpretation. The americium oxidation states were monitored by UV−vis absorption spectroscopy. Data were collected using an Agilent Cary 6000i UV−vis−NIR spectrophotometer at ambient temperature (20 ± 1 °C) in a 10 mm cuvette. A typical scan range was 1050−450 nm (0.5 nm interval; 2.0 SBW, spectral band width) to include the 996, 718, and 503 nm peaks that correspond to Am(VI), Am(V), and Am(III), respectively. Americium absorption data were baseline-corrected and converted to concentrations using the known molar extinction coefficients for Am(III), and extinction coefficient values determined by sensitivity analysis for Am(V) and Am(VI) (see the Results and Discussion in the text) using the Beer−Lambert law. 2.4. Multiscale Modeling. A combination of stochastic27−29 and deterministic methods have been used to model the complex behavior of the aforementioned americium−nitric acid systems. This approach has been successfully used to evaluate the γ-radiolysis of aqueous nitrate,30 nitric acid,31 and neptunium−nitric acid systems.25,26 The chemical-reaction set used in these calculations incorporated supplementary americium reactions given in S1 Table 1 in the Supporting Information.

density has always been the basis for separating U, Np, and Pu from the fission products since ligand-specific actinide complexation strength follows An(IV) > An(VI) ≥ An(III) > An(V).8 For example, oxidation-state adjustments are currently used to minimize or maximize the extraction of Pu in the PUREX (Plutonium Uranium Reduction EXtraction) separation schemes used worldwide.9−12 Recent studies have demonstrated the 5f-element-group coextraction of U(VI), Pu(VI), Np(VI), and Am(VI) using numerous extractants.13−15 These studies demonstrate a simplified pathway to separate americium from the trivalent lanthanides and curium, widely regarded as one of the most challenging separations in radiochemistry. The first accounts of americium oxidation were reported by Asprey, Stephanou, and Penneman in 1951.16,17 In these studies, Am(III) was completely oxidized to Am(VI) using ammonium peroxydisulfate in nitric or perchloric acids. Shortly after, Werner and Perlman isolated Am(V) from carbonate media using the oxidant sodium hypochlorite at high pH.18 These initial reports were qualitative in nature, describing the UV−vis absorbance peaks of the Am(V) and Am(VI) species and the linear reduction of Am(VI) to Am(V) over time. These early reports also suggested the Am(V) and Am(VI) species were analogous to other high-valent actinide elements (U, Np, and Pu) and were present as actinyl ions in solution. A number of studies were conducted in the following decade to characterize the newly discovered higher-valent americium species and to isolate the unobserved, but proposed, Am(IV) species, eventually yielding what are now accepted as the formal oxidation potentials for Am0 → Am(III) → Am(IV) → Am(V) → Am(VI) to be +2.07, −2.62, −0.84, and −1.60 V, respectively, in 1.0 M perchloric acid.7 The so-called autoreduction of Am(VI) was reported to occur via an autocatalytic mechanism where americium reacts with its own radiolysis products formed by α-decay.16,17,19−24 The varied experimental procedures and multiple aqueous media investigated previously have resulted in the report of inconsistent reduction mechanisms and rate constants. Given this group’s interest in the chemistry of Am(VI), its autoreduction kinetics have been reinvestigated. The oxidation experiments were conducted in aqueous nitric acid using 243Am. Nitric acid is itself redox-active under radiolysis. Therefore, the measured Am(VI) reduction kinetics were compared to modeled Am(VI) reduction kinetics, using multiscale modeling techniques that have previously been applied to understanding the radiolytic reduction of Np(VI) in nitric acid.25,26

3. RESULTS AND DISCUSSION 3.1. UV−Vis Absorption Spectroscopy. The UV−vis absorption spectra of a mixed Am(III), Am(V), and Am(VI) solution containing 2.0 mM total americium in 3.0 M HNO3 are shown in Figure 1. A series of these spectra were collected over a 24 h period, and the decrease in Am(VI) at 666 and 996 nm, the increase in Am(V) at 513 and 718 nm, and the stable

2. EXPERIMENTAL SECTION 2.1. Reagents. Weighed amounts of americium oxide (99.9% Am2O3, 0.1% 241Am2O3) from stock on-hand at Idaho National Laboratory were dissolved in 3.5 mL of 0.5 M HNO3 in a radiological glovebox to produce stock solutions of ≈2.85 mg mL−1. Aliquots of the stocks were diluted with standardized HNO3 (≥99.999% trace metal basis, Sigma-Aldrich) and 18 MΩ deionized H2O to produce a series of working solutions with concentrations of 1.0, 2.0, or 4.0 mM americium in 1.0, 3.0, or 6.5 M HNO3. The solutions were assayed by α-spectroscopy and determined to be 0.98 mM ± 4.9%, 1.95 mM ± 2.9%, and 4.01 mM ± 3.3% for the nominal 1.0, 2.0, and 4.0 mM solutions. 2.2. Americium(III) Oxidation Procedure. The oxidant sodium bismuthate (NaBiO3 , 93% ACS grade) was purchased from Chemsavers. Oxidation of Am(III) was achieved by combining 1.5 mL of an Am(III) stock solution and 60 mg solid NaBiO3 mL−1. Americium solutions of 6.5 M HNO3 required 2 h of vigorous shaking, 243

Figure 1. UV−vis absorption spectrum highlighting the multiple Am oxidation-state changes as monitored by the indicated wavelengths. The aqueous solution contained 2.0 mM total Am in 3.0 M HNO3 and was precontacted with sodium bismuthate for 24 h. The arrows indicate the change in absorption peak growth or decay. 8296

DOI: 10.1021/acs.inorgchem.7b00990 Inorg. Chem. 2017, 56, 8295−8301

Article

Inorganic Chemistry

Am(III) concentration over the entire 24 h. The slopes of these curves represent the rates of decrease in Am(VI) and increase in Am(V): −0.0132 ± 0.0023 and +0.0136 ± 0.0001 mM h−1, respectively. The mass balance for [Am]total of 1.97 mM is in excellent agreement with the α-spectroscopy assay result 1.95 mM ± 2%. This total americium concentration value may be used to calculate first-order rate constants with respect to the total americium concentration: 0.0067 h−1 for Am(VI) and 0.0069 h−1 for Am(V). Since the zero-order rate constants obtained for Am(VI) reduction are dependent upon the Am concentration, the initial slope values were divided by the total Am concentration (assayed by α-spectroscopy), to give corresponding first-order rate constants. This allowed for a better comparison of these rate constants under the various conditions of this work, as well as with previous literature ligand-degradation values. The Am(III) data may be fitted with a line of positive slope; however, the resulting rate constants are two orders of magnitude smaller than those of Am(V) and Am(VI). Thus, the agreement between these rates indicates that Am(VI) undergoes a 1-electron reduction to Am(V), and the nearly constant Am(III) concentration indicates that Am(V) is stable over the 24 h period of measurement. Any proposed competing reactions, such as a 2-electron reduction of Am(VI) to Am(IV) followed by an immediate reduction to Am(III), or Am(V) disproportionation, either do not occur or are only marginally significant since the Am(III) concentration remains nearly unchanged. Analogous data were collected over the range 1.0−6.5 M HNO3, with 1.0−4.0 mM total americium concentrations. The resulting rate constants are shown in Table 2. The corresponding plots and the associated rates of Am(VI) reduction and Am(V) growth are given in SI Figures 1−6. For each experiment, the rates for Am(VI) reduction and Am(V) growth agree, as was shown in Figure 2. The rate of Am(VI) reduction increases with total americium concentration, as shown for the 3.0 M HNO3 experiments in Figure 3 (panel A), where the rate of reduction of Am(VI) approximately doubles for each factor-of-2 increase in total americium concentration. This is consistent with the reduction being due to the reaction of hexavalent americium with total americium α-radiolysis products. Analogous data were obtained for 6.5 M HNO3, and are shown in SI Figure 7. The data in Figure 3B show that the rate of Am(VI) autoreduction varies with nitric acid concentration for a given total americium concentration. The highest rates are observed in 3.0 M HNO3, with lower values in 1.0 and 6.5 M HNO3 (Table 3). These results are not expected since the highest radiolytic production of HNO2 is at the highest nitric acid concentration.25,26 This may indicate competition for reaction with Am(VI) by the products of both water and HNO3 radiolysis, with the intermediate acid concentration being the matrix in which the highest concentrations of both types of reactive species occur. This is discussed in more detail later. For comparison, Table 2 also contains kinetic data for the reduction of Am(VI) when oxidized using the alternative oxidant copper(III) periodate (see SI Figure 8).35 The autoreduction rate constant reported for Cu(III)-oxidized Am(VI) is 0.0073 ± 0.0004 h−1 in 3.0 M HNO3, and shows no statistical difference (at 1 σ error) when compared to the averaged values for Am(VI) reduction (0.0074 ± 0.0004 h−1 in 3.0 M HNO3) from bismuthate oxidation (Table 2). This also confirms that any residual bismuthate does not interfere with

to slightly increasing absorbance of Am(III) at 503 and 810 nm are shown. 3.2. Respective Molar Extinction Coefficients for the Americium Species. All three americium oxidation states (III, V, and VI) were observed at their absorbance maxima of 503 [Am(III)], 718 [Am(V)], and 996 nm [Am(VI)]. The molar extinction coefficients of penta- and hexavalent americium are not well-characterized, especially for the HNO3 concentrations investigated in this work. For Am(III), the extinction coefficient for the 503 nm peak was extrapolated from the values provided over the range 0.1−5.9 M HNO3 reported by Zalupski et al.32 Literature values of the molar extinction coefficient for Am(V), ε718, vary from 59.3 to 66 M−1 cm−1 in dilute HClO4 and/or H2SO4.33 No molar extinction coefficients were available in the literature for the 718 nm absorption peak in HNO3 media. The Am(VI) ε996 values vary from 63.8 to 83.3 M−1 cm−1 in varying concentrations of HClO4,33 while the review by Schulz34 reports a value of approximately 70 M−1 cm−1 in 3.0−6.0 M HNO3. With initial values within these ranges, the extinction coefficients for Am(V) and Am(VI) were varied until an acceptable mass balance for the total americium concentration was obtained for each experiment. The molar extinction coefficients obtained in this work are summarized for all experiments in Table 1. Table 1. Summary of Molar Extinction Coefficients (ε) Used To Calculate the Concentrations of Each Americium Oxidation State as a Function of Nitric Acid and Total Americium Concentration molar extinction coefficient, M−1 cm−1 [Am], mM

oxidant

4.0

NaBiO3

1.0 2.0 4.0 1.0 average

NaBiO3 NaBiO3 NaBiO3 Cu(III)

1.0 2.0 4.0 4.0

NaBiO3 NaBiO3 NaBiO3 NaBiO3/ dodecane

average

ε503 Am(III) 1.0 M HNO3 391 3.0 M HNO3 356 356 356 356 6.5 M HNO3 280 280 280 280

ε718 Am(V)

ε996 Am(VI)

37.5

78.0

48.8 43.0 38.2 45.0 43.8 ± 4.41

92.6 95.0 86.5 91.0 91.3 ± 3.58

35.7 37.5 37.0 37.0

81.0 81.0 80.0 80.0

36.8 ± 0.77

80.5 ± 0.58

Figure 2 shows the baseline-corrected spectra and the corresponding Am(VI) kinetic plots for 2.0 mM total americium in 3.0 M HNO3. For this and all experiments, the first 1 h of data collection was truncated, because it did not always observe linearity. This was also observed by Hall and Markin23 and may indicate trace reducing agents present as impurities. Once these are consumed, the reduction of Am(VI) is attributed to self-radiolysis only. The corresponding americium spectra are shown in panels A, B, and C of Figure 2, which provide the basis for the concentration versus time plot in panel D. The linear plots of panel D show zero-order kinetics, consistent with autoreduction. The data show both the reduction of Am(VI) and the ingrowth of Am(V) to be linear, with little or no change in 8297

DOI: 10.1021/acs.inorgchem.7b00990 Inorg. Chem. 2017, 56, 8295−8301

Article

Inorganic Chemistry

Figure 2. Panels A, B, and C show the baseline-corrected absorption peaks for 996 nm [Am(VI)], 718 nm [Am(V)], and 503 nm [Am(III)] used to calculate concentrations of the Am species in solution. Panel D shows the autoreduction of Am(VI) (red ■) in 3.0 M HNO3, for a total Am concentration of nominally 2.0 mM, and the simultaneous change in concentration for Am(V) (blue ●) and Am(III) (brown ◆). The mass balance (teal ▲) for total americium is 1.970 ± 0.007 mM. The following molar extinction coefficients were used: ε503 = 356; ε718 = 43.0; ε996 = 95.0 M−1 cm−1. The best-fit lines in panel D were obtained using least-squares linear regression equations.

0.031 ± 0.001 h−1 in 0.2 M HClO4. Those studies were conducted using 241Am with a half-life of 432.7 years. The 243 Am used in the current work has a half-life of 7.37 × 103 years, but a similar α-decay energy of 5.4 MeV versus 5.2 MeV. Consequently, the yield of radiolysis products capable of reducing Am(VI) is expected to be greater for 241Am by the ratio of these half-lives, which should result in a rate constant 17 times faster for 241Am. The rate constant reported by Hall and Markin23 is only 6.5 times faster than the slowest rate constant reported here, indicating that the irradiated HClO4 medium does not have a radiation chemical yield of reducing agents as high as that of HNO3. This strongly suggests that HNO2 is indeed an important reducing agent in addition to the H2O2 produced by the α-radiolysis of water. The only data available under comparable conditions to those of the current work is by Zaitsev,36 as reviewed by Schulz34 who reported a rate constant [241Am(VI)] of 0.087 h−1 for 3.0−6.0 M HNO3, about 12 times faster than the highest value reported here. In an experiment to simulate solvent extraction conditions similar to the fuel cycle, an equal volume of dodecane was layered over the aqueous phase during the 24 h UV−vis data collection. Data for the reduction of Am(VI) in 4.0 mM/6.5 M HNO3 in constant contact with dodecane is shown in Figure 4, with the rate constants also summarized in Table 2. The rate constant for the growth of Am(V) was 0.0056 ± 0.0004 h−1, the same as that which was found for this acid concentration in the absence of dodecane (average 0.0054 ± 0.0003 h−1 in Table 2), suggesting that the rate of Am(VI) autoreduction was

Table 2. Summary of Am(VI) Autoreduction and Am(V) Growth First-Order Rate Constants rate constants, h−1 [Am], mM

Am(VI)

4.0

0.0049 ±

1.0 2.0 4.0 1.0a average

0.0081 0.0067 0.0075 0.0073

1.0 2.0 4.0 average 4.0b

± ± ± ±

Am(V)

1.0 M HNO3 0.0003 0.0046 ± 0.0003 3.0 M HNO3 0.0004 0.0079 ± 0.0004 0.0002 0.0069 ± 0.0001 0.0004 0.0077 ± 0.0004 0.0004 0.0080 ± 0.0004

6.5 M HNO3 0.0062 ± 0.0003 0.0050 ± 0.0003 0.0054 ± 0.0003 0.0051 ± 0.0003 0.0047 ± 0.0003 0.0045 ± 0.0003 0.0128 ± 0.0007

average 0.0048 ± 0.0003 0.0080 0.0068 0.0076 0.0077 0.0075

± ± ± ± ±

0.0004 0.0003 0.0004 0.0004 0.0005

0.0056 0.0053 0.0046 0.0054

± ± ± ±

0.0003 0.0003 0.0003 0.0003

0.0056 ± 0.0004

a

With Cu(III)-oxidized Am(VI). bReduction of Am(VI) in constant contact with dodecane.

the autoreduction process being measured, at least after the initial 1 h when data collection was initiated. The rate constants given in Table 2 are difficult to compare to those of previous works given the disparities in the matrix, and the isotope of americium used. For example, Hall and Markin23 reported an Am(VI) autoreduction rate constant of 8298

DOI: 10.1021/acs.inorgchem.7b00990 Inorg. Chem. 2017, 56, 8295−8301

Article

Inorganic Chemistry

Figure 4. Autoreduction of Am(VI) (red ■) in 6.5 M HNO3 in constant contact with an equal volume of dodecane, for a total Am concentration of nominally 4.0 mM, and the simultaneous change in concentration for Am(V) (blue ●) and Am(III) (brown ◆). The mass balance (teal ▲) for total americium is 3.93 ± 0.05 mM. The following extinction coefficients were used: ε503 = 280; ε718 = 37.0; ε996 = 80.0 M−1 cm−1. The best-fit lines were obtained using least-squares linear regression equations.

included concurrent 1-electron and 2-electron Am(VI) reduction pathways. In that report, americium speciation remained static after all Am(VI) was reduced, indicating that Am(V) is not as prone to autoreduction, consistent with the fact that Am(V) is a weak oxidant (Ered = 0.84 V versus SCE).7 3.3. Mechanistic Considerations. The main reduction pathway for Am(VI) in aqueous HNO3 might be assumed to be the reaction with radiolytically produced H2O2 (eq 1),34 because of it being a high-yield product of the α-radiolysis of water.38

Figure 3. (A) Rates of change describing the production of Am(V) (blue ●) and loss of Am(VI) (red ■) as a function of total americium concentration in 3.0 M HNO3. The best-fit lines were obtained using least-squares linear regression equations. (B) Rates of change describing Am(V) growth (blue ●) and Am(VI) reduction (red ■) for 4.0 mM americium as a function of nitric acid concentration. The lines in panel B are presented only to guide the eyes and do not represent a best fit.

Table 3. Comparison of Calculated Radiation Track Yields and Empirical Direct Effect Yields for the Key Reducing Species at 10 μs for the α-Radiolysis of Aerated Aqueous HNO3 Solutions for a 5.4 MeV α-Particle

AmO2 2 + + H 2O2 → AmO2+ + •HO2 + Haq +

However, the α-radiolysis of nitric acid yields comparable amounts of HNO2, which was shown to be an important reducing species in the neptunium−nitric acid system.25,26 The significance of HNO2 formation is that it is also a predominant Am(VI)-reducing species and that it reacts with H2O2 (eq 3).33 Thus, the equilibrium between these radiolytically produced species determines the final concentration of species available to reduce Am(VI).25

radiolytic yields (species 100 eV−1)

a

species

1.0 M HNO3

3.0 M HNO3

6.5 M HNO3

H2O2 NO32− HNO3− HNO2,direct‑effecta reducing capability

1.37 2.61 0.34 0.23 4.55

1.35 1.57 1.3 0.64 4.86

1.23 1.6 1.39 1.37 5.59

(1)

AmO2 2 + + HNO2 → AmO2+ + NO2 + Haq +

Derived from ref 40.

HNO2 + H 2O2 ⇌ ONOOH + H 2O

(2)

k 3f = 7.17 × 10 M s−1 5

k 3b = 300 M s−1

unchanged. However, the 2-electron reduction is apparently no longer insignificant with a rate constant for Am(III) production of 0.0053 ± 0.0004 h−1. Trivalent americium is produced by the instantaneous reduction of the Am(IV) [(Ered = 2.6 V versus SCE]7 produced by the 2-electron reduction of hexavalent americium. The combination of these two pathways in the presence of the dodecane explains the higher Am(VI) reduction rate constant with its value of 0.0128 ± 0.0007 h−1. Similar data to those collected here in the presence of dodecane were reported by Dares et al.37 for the reduction of Am(VI) in 0.1 M HNO3 solution where the high-valent americium was prepared by electrochemical oxidation using surface-functionalized electrodes. These electrodes contained organic complexing agents. The mechanism they proposed also

(3)

The data show that the rate constants for Am(VI) autoreduction vary as a function of HNO3 concentration (Figure 3 and SI Figure 7). The highest rate constant is found in 3.0 M HNO3, with lower values in 1.0 and 6.5 M HNO3. For a start in understanding the observed kinetic trend with solution acidity, the changes in both radiation track chemistry and bulk homogeneous chemistry must be fully considered. 3.4. Radiation Track Chemistry. As the concentration of HNO3 (HNO3 ⇌ NO3− + Haq+, pKa ≈ −1.37)39 increases, the scavenging capacities (ks = k × [scavenger]) of the nitrate anion (NO3−) for the hydrated electron [eaq−], its precursor (epre−), and hydrogen atom (H•) increase proportionally, generally 8299

DOI: 10.1021/acs.inorgchem.7b00990 Inorg. Chem. 2017, 56, 8295−8301

Article

Inorganic Chemistry

for reaction with Am(VI) than at 1.0 or 6.5 M HNO3. As such, this modeling demonstrates that the autoreduction of hexavalent americium in nitric acid solutions is a consequence of the reactions of HNO2, and to a lesser extent H2O2.

leading to progressively higher radiolytic yields of reduced NO3− intermediates (NO3•2− and HNO3•−), as given by reactions expressed in eqs 4−7. H 2O ⇝ eaq −, Haq +, H•, OH•, H 2 , H 2O2

NO3− + e pre− → NO3•2 −

(4)

4. CONCLUSIONS The rate of hexavalent americium autoreduction has been measured across a range of nitric acid concentrations. These rates are reported for Am(VI) autoreduction using 243Am in a nitric acid solution free of any potential trace reductants. The reduction follows zero-order kinetics with respect to [Am(VI)] but first-order kinetics with respect to total americium concentration, because of the reaction with the radiolysis products of americium decay. The empirical first-order rate constants are given, and compared to those of the limited previous work using 241Am. Multiscale radiation chemical modeling including the α-radiolysis yields for HNO2 and H2O2 in irradiated aqueous nitric acid agreed with the empirically modeled rate of autoreduction at various acid concentrations, strongly suggesting the proposed radiolytic mechanism. Americium speciation in these experiments was observed by UV−vis spectroscopy, necessitating the determination of the unknown molar extinction coefficients for the Am(V) 718 and Am(VI) 996 nm peaks. These were obtained by iteration from initial reasonable values to obtain the total americium mass balance with the known total americium concentrations. The new extinction coefficients reported here are within a general uncertainty of less than ±4% for 6.5 M HNO3, and ±10% for the 3.0 M HNO3 conditions. Multiscale modeling demonstrated that HNO2 is the predominant reducing species under self-radiolysis conditions, as H2O2 possesses a lower radiolytic yield and also reacts with HNO2. Furthermore, the trends exhibited by the empirical Am(VI) autoreduction rate constants are a consequence of bulk homogeneous effects affecting the available proportions of HNO2 and H2O2 such that Am(VI) is least stable in 3.0 M HNO3, relative to 1.0 and 6.5 M HNO3. Understanding the autoreduction of Am(VI) in HNO3 solutions yields significant insight into the effects of α-radiolysis on potential americyl reprocessing flowsheets.

k5 = 1 × 1013 M−1 s−1 (5)

NO3− + eaq − → NO3•2 −

k6 = 9.7 × 109 M−1 s−1 (6)

NO3− + H• → HNO3•−

k 7 = 1.0 × 107 M−1 s−1 (7)

NO3−

These reduced intermediates are precursors that will form HNO2, and thus their radiolytic yields, in conjunction with those for H2O2 from water radiolysis (eq 4) and HNO2 from direct effect formation (HNO3 ⇝ HNO2 + O), are an approximation of the reducing capability of the medium with respect to Am(VI). Table 3 gives the radiation track yields used by the multiscale modeling approach29 for a 5.4 MeV α-particle as a function of HNO3 concentration. The yields of Am(VI)-reducing species show a significant increase with increasing HNO3 concentration, which is expected from scavenging capacity and energy deposition partitioning arguments. On the basis of changes in track chemistry alone, it would be expected for Am(VI) to be progressively less stable with increasing HNO3 concentration because of the higher reducing capability. However, this was not observed, and indicates that the trend exhibited by the empirical Am(VI) autoreduction rate coefficients is predominantly a product of bulk homogeneous chemistry. 3.5. Bulk Homogeneous Chemistry. The concentration of radiolytically produced HNO2 increases with HNO3 concentration, which is evident from the radiolytic yields given in Table 3. This leads to increased competition for HNO2 between Am(VI) and H2O2. The accumulated product concentrations for the reaction of HNO2 and H2O2 as a function of nitric acid concentration are shown in Figure 5 which demonstrates that this reaction is least efficient at 3.0 M HNO3. Consequently, in 3.0 M HNO3, more HNO2 is available

■

ASSOCIATED CONTENT

* Supporting Information S

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.inorgchem.7b00990. Figures of americium concentration over time, figure of the rate of change versus americium concentration, and table of americium reactions and rate coefficients used in the multiscale modeling approach (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Travis S. Grimes: 0000-0003-2751-0492 Gregory P. Horne: 0000-0003-0596-0660

Figure 5. Accumulated product concentrations for the reaction of HNO2 + H2O2 from 2.0 mM Am−HNO3 solutions predicted by multiscale modeling calculations: 1.0 (black ), 3.0 (red −−−), and 6.5 M HNO3 (blue −·−).

Notes

The authors declare no competing financial interest. 8300

DOI: 10.1021/acs.inorgchem.7b00990 Inorg. Chem. 2017, 56, 8295−8301

Article

Inorganic Chemistry

■

(19) Gunn, S. R.; Cunningham, B. B. The Heats of Formation of AmO2+ (aq) and AmO22+ (aq) in 1 M HClO4. J. Am. Chem. Soc. 1957, 79 (7), 1563−1565. (20) Penneman, R. A.; Asprey, L. B. A Review of Americium and Curium Chemistry, 1st International Conference on Peaceful Uses of Atomic Energy, Geneva; 1956, Vol. 7; pp 355−362. (21) Coleman, J. S. The Kinetics of the Disproportionation of Americium(V). Inorg. Chem. 1963, 2 (1), 53−57. (22) Hall, G. R.; Herniman, P. D. The Separation and Purification of Americium-241 and the Absorption Spectra of Tervalent and Quinquevalent Americium Solutions. J. Chem. Soc. 1954, 2214−2221. (23) Hall, G. R.; Markin, T. L. The Self-Reduction of Americium(V) and (VI) and the Disproportionation of Americium(V) in Aqueous Solution. J. Inorg. Nucl. Chem. 1957, 4, 296−303. (24) Keenan, T. K. Americium and Curium. J. Chem. Educ. 1959, 36 (1), 27−31. (25) Mincher, B. J.; Precek, M.; Mezyk, S. P.; Elias, G.; Martin, L. R.; Paulenova, A. The Redox Chemistry of Neptunium in γ-irradiated Aqueous Nitric Acid. Radiochim. Acta. 2013, 101, 259−265. (26) Horne, G. P.; Grimes, T. S.; Mincher, B. J.; Mezyk, S. P. Reevaluation of Neptunium-nitric Acid Radiation Chemistry by MultiSacale Modeling. J. Phys. Chem. B 2016, 120 (49), 12643−12649. (27) Pimblott, S. M.; LaVerne, J. A.; Mozumder, A. Monte Carlo Simulation of Range and Energy Deposition by Electrons in Gaseous and Liquid Water. J. Phys. Chem. 1996, 100, 8595−8606. (28) Pimblott, S. M.; LaVerne, J. A. Effects of Track Structure on the Ion Radiolysis of the Fricke Dosimeter. J. Phys. Chem. A 2002, 106, 9420−9427. (29) Clifford, P.; Green, N. J. B.; Oldfield, M. J.; Pilling, M. J.; Pimblott, S. M. Stochastic Models of Multi-Species Kinetics in Radiation-Induced Spurs. J. Chem. Soc., Faraday Trans. 1 1986, 82, 2673−2689. (30) Horne, G. P.; Donoclift, T. A.; Sims, H. E.; Orr, R. M.; Pimblott, S. M. Multi-Scale Modelling of the Gamma Radiolysis of Nitrate Solutions. J. Phys. Chem. B 2016, 120, 11781−11789. (31) Garaix, G.; Horne, G. P.; Venault, L.; Moisy, P.; Pimblott, S. M.; Marignier, J.-L.; Mostafavi, M. Decay Mechanism of NO3• Radical in Highly Concentrated Nitrate and Nitric Acidic Solutions in the Absence and Presence of Hydrazine. J. Phys. Chem. B 2016, 120, 5008−5014. (32) Zalupski, P. R.; Grimes, T. S.; Heathman, C. R. Spectroscopy. Manuscript in preparation. (33) Runde, W. H.; Mincher, B. J. Higher Oxidation States of Americium: Preparation, Characterization and Use for Separations. Chem. Rev. 2011, 111, 5723−5741. (34) Schulz, W. W. The Chemistry of Americium; DOE Technical Information Center: Oak Ridge, TN, 1976. (35) Sinkov, S. I.; Lumetta, G. J. Americium(III) Oxidation by Copper(III) Periodate in Nitric Acid Solution as Compared with the Action of Bi(V) compounds of Sodium, Lithium, and Potassium. Radiochim. Acta 2015, 103 (8), 541−552. (36) Zaitsev, A. A.; Kosyakov, V. N.; Rykov, A. G.; Sobolov, Y. P.; Yakolev, G. N. Radiolytic Reduction of Am(VI) and Am(V). At. Energy (N. Y., NY, U. S.) 1960, 7, 562−569. (37) Dares, C. J.; Lapides, A. M.; Mincher, B. J.; Meyer, T. J. Electrochemical Oxidation of 243Am(III) in Nitric Acid by a Terpyridyl-derivatized Electrode. Science 2015, 350 (6261), 652−655. (38) LaVerne, J. A. Radiation chemical effects of heavy ions. In Charged Particle and Photon Interactions with Matter; Mozumder, A., Hatano, Y., Eds.; Marcel Dekker: New York, 2004; pp 403−429. (39) Davis, W.; De Bruin, H. J. New Activity Coefficients of 0 − 100 Per Cent Aqueous Nitric Acid. J. Inorg. Nucl. Chem. 1964, 26, 1069− 1083. (40) Jiang, P. Y.; Nagaishi, R.; Yotsuyanagi, T.; Katsumura, Y.; Ishigure, K. Gamma Radiolysis Study of Concentrated Nitric Acid Solutions. J. Chem. Soc., Faraday Trans. 1994, 90, 93−95.

ACKNOWLEDGMENTS T.S.G. and B.J.M. were funded by the US DOE Office of Nuclear Energy FCR&D Sigma Team for Advanced Actinide Recycle under Idaho Operations Contract DEAC0705ID14517; G.P.H. and S.P.M. were funded by the US DOE Assistant Secretary for NE, under the FCR&D Radiation Chemistry program: DOE-Idaho Operations Office Contract DE-AC07-05ID14517 and Nuclear Energy University Partnership grant DE-NE0008406. C.J.D. was funded by DOE-NEUP under contract DE-NE0008539. The authors would like to thank Professor Jen Braley and Kevin McCann (CSM) for providing the copper(III) periodate used in this study.

■

REFERENCES

(1) Jha, M. K.; Kumar, V.; Singh, R. J. Review of Hydrometallurgical Recovery of Zinc from Industrial Wastes. Resour. Conserv. Recycl. 2001, 33, 1−22. (2) Habashi, G. A. Short History of Hydrometallurgy. Hydrometallurgy 2005, 79, 15−22. (3) Rudnik, E.; Nikiel, M. Hydrometallurgical Recovery of Cadmium and Nickel from Spent Ni-Cd Batteries. Hydrometallurgy 2007, 89, 61−71. (4) Denning, R. G.; Norris, J. O. W.; Short, I. G.; Snellgrove, T. R.; Woodwark, D. R. In Lanthanide and Actinide Chemistry and Spectroscopy. Edelstein, N. M., Ed.; ACS Symposium Series 131; American Chemical Society: Washington, DC, 1980; pp 313−330. (5) Neuefeind, J.; Soderholm, L.; Skanthakumar, S. Experimental Coordination Environment of Uranyl(VI) in Aqueous Solution. J. Phys. Chem. A 2004, 108, 2733−2739. (6) Denning, R. G. Electronic Structure and Bonding in Actinyl Ions and their Analogs. J. Phys. Chem. A 2007, 111, 4125−4143. (7) Bard, A. J.; Parsons, R.; Jordan, J. Standard Potentials in Aqueous Solution; Marcel Dekker, Inc.: New York, 1985. (8) Maher, K.; Bargar, J. R.; Brown, G. E., Jr. Environmental Speciation of Actinides. Inorg. Chem. 2013, 52, 3510−3532. (9) Choppin, G. R.; Rydberg, J.-O. Treatment of spent nuclear fuel. In Nuclear Chemistry: Theory and Applications, 1st ed.; Pergamon Press Inc.: New York, 1980; pp 502−559. (10) Benedict, M.; Pigford, T. H.; Levi, H. W. Nuclear Chemical Engineering, 2nd ed.; McGraw-Hill: New York, 1981. (11) Choppin, G. R.; Liljenzin, J.-O.; Rydberg, J. Nuclear fuel cycle. In Radiochemistry and Nuclear Chemistry, 3rd ed.; ButterworthHeinemann: Woburn, MA, 2002; pp 582−641. (12) Todd, T. A.; Wigeland, R. A. Advanced separation technologies for processing used nuclear fuel and the potential benefits to a geologic repository. In Separations for the Nuclear Fuel Cycle in the 21st Century; Lumetta, G. L., Nash, K. L., Clark, S. B., Friese, J. I., Eds.; ACS Symposium Series 933; American Chemical Society: Washington DC, 2006; pp 41−55. (13) Mincher, B. J.; Martin, L. R.; Schmitt, N. C. Tributylphosphate Extraction Behavior of Bismuthate-Oxidized Americium. Inorg. Chem. 2008, 47, 6984−6989. (14) Mincher, B. J.; Martin, L. R.; Schmitt, N. C. Diamylamylphosphonate Solvent Extraction of Am(VI) from Nuclear Fuel Raffinate Simulant Solution. Solvent Extr. Ion Exch. 2012, 30, 445−456. (15) McCann, K.; Mincher, B. J.; Schmitt, N. C.; Braley, J. C. Hexavalent actinide extraction using N,N-dialkyl Amides. Ind. Eng. Chem. Res. 2017, 56, 6515. (16) Asprey, L. B.; Stephanou, S. E.; Penneman, R. A. A New Valence State of Americium, Am(VI). J. Am. Chem. Soc. 1950, 72 (3), 1425− 1426. (17) Asprey, L. B.; Stephanou, S. E.; Penneman, R. A. Hexavalent Americium. J. Am. Chem. Soc. 1951, 73 (12), 5715−5717. (18) Werner, L. B.; Perlman, I. The Pentavalent State of Americium. J. Am. Chem. Soc. 1951, 73 (1), 495−496. 8301

DOI: 10.1021/acs.inorgchem.7b00990 Inorg. Chem. 2017, 56, 8295−8301